Georeferencing Of Museum Collections: A Review Of Problems And Automated Tools, And The Methodology Developed By The Mountain And Plains Spatio-Temporal Database-Informatics Initiative (Mapstedi)

The vast majority of locality descriptions associated with biological specimens housed in natural history museums lack the geographic coordinates required for computer-based geographic analyses. Assigning such coordinates to existing specimen records is a process called retrospective georeferencing. The georeferencing of biological collections makes those collections more valuable by allowing them to be used in spatially explicit biodiversity analyses. Here we review some of the most common problems inherent to the retrospective georeferencing of biological collections. We then attempt to classify the most common types of locality descriptions according to our own rule-application for georeferencing, which was developed as part of a larger funded effort to create an online mapping and biodiversity analysis portal for the North-central Rocky Mountains and adjacent plains. As a means of comparison with our manual computer-assisted georeferencing method, we evaluate four currently available automated georeferencing tools. We argue that these automated tools are less accurate and precise, and less time efficient, than the three computer-assisted manual protocols for natural history museums: MaNIS, MaPSTeDI, and INRAM. Nevertheless, it is clear that these automated tools represent a critical first-step in the development of accurate, precise and efficient tools which will ultimately replace or at least enhance existing manual protocols. In order to facilitate the development of better georeferencing technologies, it is critical that workers involved establish a consensus set of locality types and rules for georeferencing them which will pave the way for the next generation of semi-or fully automated tools. A collaboratively built next generation georeferencing tool should be Internet-based.